Abstract: FR-PO036
Data-Driven Disease Detection: A Learning Health System
Session Information
- AI, Digital Health, Data Science - II
November 03, 2023 | Location: Exhibit Hall, Pennsylvania Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Augmented Intelligence, Digital Health, and Data Science
- 300 Augmented Intelligence, Digital Health, and Data Science
Authors
- Butler, Grainne H., Centre for Health Analytics, Royal Children's Hospital Melbourne, Melbourne, Victoria, Australia
- Shanks, Josiah, The Royal Children's Hospital Melbourne, Parkville, Victoria, Australia
- East, Casie, Centre for Health Analytics, Royal Children's Hospital Melbourne, Melbourne, Victoria, Australia
- Satkumaran, Saravanan, Centre for Health Analytics, Royal Children's Hospital Melbourne, Melbourne, Victoria, Australia
- Quinlan, Catherine, The Royal Children's Hospital Melbourne, Parkville, Victoria, Australia
Background
Achieving pre-clinical diagnosis in time to change the course of disease is the promise of data analytics and genomic medicine.Our Kidney Genomics Clinic(KGC) showed a high diagnostic yield for Alport syndrome(AS) in a pediatric cohort with persistent microscopic haematuria.Early diagnosis of AS in childhood offers an opportunity to start treatment and delay progression.Large studies of variant databases have shown heterozygous pathogenic variants in COL4A3/COL4A4 are present in 1 in 106 and pathogenic variants in COL4A5 in 1 in 2320.
Despite guidelines and education,referral rates remained static and follow up rates were low.We created a digital health solution using data analytics of the electronic medical record(EMR) to identify children with an incidental finding of microscopic haematuria and offer early genomic diagnosis to those at risk of AS.
Methods
EMR databases were interrogated using Microsoft SQL to identify patients with incidental findings of microscopic haematuria and no prior or subsequent results below the threshold(20 RCC).Patients were stratified according to levels of haematuria and those with pyuria excluded. Patients were contacted with a letter outlining the need for repeat testing utilising bulk communication and ordering functions within the EMR.All those with persistent microscopic haematuria were offered genomic sequencing through the KGC.We developed a suite of web-based resources to enhance recruitment and webpage analytics were monitored for views and engagement.As a learning health system,we sought to integrate feedback and lessons learned to iteratively improve each round of patient contact.
Results
Figure 1
Conclusion
Direct patient contact through the EMR is feasible and acceptable demonstrating how digital health strategies can be employed to solve problems in healthcare.Response rates using postal contact are poor;alternate modes of communication may improve uptake.Lessons learned from this will be utilized to develop a prospective strategy automating the process for recall of future findings of microscopic haematuria.